Image processing apparatus and method

Abstract

Upon gradation converting input m-bit image information into n-bit image information (m and n are integers; m>n), random numbers, the maximum values of which are changed in accordance with the input values of the m-bit image information, are generated at periods of two or more pixels, and the random numbers are selectively output while switching their signs to positive/negative within the period of two or more pixels. The output signal values and the input values of the m-bit image information are added to each other, and the sum signal values are converted into n-values by a density preservation type quantization method. With this control, the problems of sweeping, texture, and the like of the conventional density preservation type quantization method can be solved without emphasizing the granularity of an image.

Claims

1. An image processing apparatus for performing gradation conversion of input m-bit image information into n-bit image information (m and n are integers and m>n), and outputting the converted image information, comprising:

random number generation means for generating a random number;

control means for switching a sign of the random number from positive to negative or from negative to positive at a predetermined period and outputting said random number; and

conversion means for converting an m-bit input value into n-bit data by quantization method of a density preservation type, on the basis of the signal value output from said control means and the input value of the m-bit image information.

2. The apparatus according to claim 1, wherein said random number generation means generates the random number in correspondence with the input m-bit image information.

3. The apparatus according to claim 1, wherein said random number generation means generates each random number at a period of not less than two pixels, and

said control means inverts the sign of the random number from positive to negative or from negative to positive within the period of not less than two pixels and selectively outputs the inverted signal value.

4. The apparatus according to claim 1, further comprising edge detection means for detecting edge information from the input image information,

wherein said random number generation means controls a value of a random number to be generated in accordance with the edge information.

5. The apparatus according to claim 1, wherein said conversion means adds the signal value from said control means and the m-bit image information, and converts the resultant sum signal value to n-bit data on the basis of a threshold value.

6. The apparatus according to claim 1, wherein said conversion means corrects a threshold value on the basis of the signal value from said control means, and converts the m-bit image information into the n-bit data on the basis of the corrected threshold value.

7. An image processing apparatus for performing gradation conversion of input m-bit image information into n-bit image information (m and n are integers and m>n), and outputting the converted image information, comprising:

random number generation means for generating a random number at a period of not less than three pixels;

control means for inverting a sign of the random number generated by said random number generation means from positive to negative or from negative to positive within the period of not less than three pixels, and selectively outputting the inverted signal value and a signal value "0"; and

conversion means for converting an m-bit input value into n-bit data on the basis of the signal value output from said control means and the input value of the m-bit image information.

8. The apparatus according to claim 7, wherein said random number generation means generates the random number in correspondence with the input m-bit image information.

9. The apparatus according to claim 7, further comprising edge detection means for detecting edge information from the input image information,

wherein said random number generation means controls a value of a random number to be generated in accordance with the edge information.

10. The apparatus according to claim 7, wherein said conversion means adds the signal value from said control means and the m-bit image information, and converts the resultant sum signal value to n-bit data on the basis of a threshold value.

11. The apparatus according to claim 7, wherein said conversion means corrects a threshold value on the basis of the signal value from said control means, and converts the m-bit image information into the n-bit data on the basis of the corrected threshold value.

12. The apparatus according to claim 7, wherein said conversion means converts the m-bit input value into n-bit data by quantization method of a density preservation type.

13. An image processing method of performing gradation conversion of input m-bit image information into n-bit image information (m and n are integers and m>n), and outputting the converted image information, comprising the steps of:

generating a random number;

controlling a switching of a sign of the random number from positive to negative or from negative to positive at a predetermined period and outputting said random number; and

converting an m-bit input value into n-bit data by quantization method of a density preservation type, on the basis of the signal value output at said control step and the input value of the m-bit image information.

14. The method according to claim 13, wherein said generating step generates the random number in correspondence with the input m-bit image information.

15. The method according to claim 13, wherein said generating step generates each random number at a period of not less than two pixels, and

said control step inverts the sign of the random number from positive to negative or from negative to positive within the period of not less than two pixels and selectively outputs the inverted signal value.

16. The method according to claim 13, further comprising a step of detecting edge information from the input image information,

wherein said generating step controls a value of a random number to be generated in accordance with the edge information.

17. The method according to claim 13, wherein said converting step adds the signal value output at said controlling step and the m-bit image information, and converts the resultant sum signal value to n-bit data on the basis of a threshold value.

18. The apparatus according to claim 13, wherein said converting step corrects a threshold value on the basis of the signal value output at said controlling step, and converts the m-bit image information into the n-bit data on the basis of the corrected threshold value.

19. An image processing method of performing gradation conversion of input m-bit image information into n-bit image information (m and n are integers and m>n), and outputting the converted image information, comprising the step of:

generating a random number at a period of not less than three pixels;

controlling an inversion of a sign of the random number generated at said generating step from positive to negative or from negative to positive within the period of not less than three pixels, and a selective output of the inverted signal value and a signal value "0"; and

converting an m-bit input value into n-bit data on the basis of the signal value output at said controlling step and the input value of the m-bit image information.

20. The method according to claim 19, wherein said generating step generates the random number in correspondence with the input m-bit image information.

21. The method according to claim 19, further comprising a step of detecting edge information from the input image information,

wherein said generating step controls a value of a random number to be generated in accordance with the edge information.

22. The method according to claim 19, wherein said converting step adds the signal value output at said controlling step and the m-bit image information, and converts the resultant sum signal value to n-bit data on the basis of a threshold value.

23. The method according to claim 19, wherein said converting step corrects a threshold value on the basis of the signal value output at said controlling step, and converts the m-bit image information into the n-bit data on the basis of the corrected threshold value.

24. The method according to claim 19, wherein said converting step converts the m-bit input value into n-bit data by quantization method of a density preservation type.

25. An image processing apparatus comprising:

image input means for inputting m-bit image information;

random number generation means for generating a random number value;

control means for randomly converting a sign of the random number value generated by said random number generation means to positive or negative and outputting the random number value; and

conversion means for converting the m-bit image information into n-bit data on the basis of the random number value output from said control means and the m-bit image information.

26. The apparatus according to claim 25, wherein said random number generation means generates the random number in correspondence with the input m-bit image information.

27. The apparatus according to claim 25, wherein said conversion means converts the m-bit input value into n-bit data by quantization method of a density preservation type.

28. An image processing method comprising the step of:

inputting m-bit image information;

generating a random number value;

controlling randomly a conversion of a sign of the random number value generated at said generating step to positive or negative and an output of the random number value; and

converting the m-bit image information into n-bit data on the basis of the random number value output at said controlling step and the m-bit image information.

29. The method according to claim 28, wherein said generating step generates the random number in correspondence with the input m-bit image information.

30. The method according to claim 28, wherein said converting step converts the m-bit input value into n-bit data by quantization method of a density preservation type.

31. An image processing apparatus comprising:

image input means for inputting m-bit image information;

random number generation means for generating positive random number values at every other pixel;

random number sign control means for randomly converting the signs of the random number values generated by said random number generation means to positive or negative;

amplitude control means for changing a ratio of an amplitude of each of the random number values output from said random number sign control means as a function of a signal value input by said image input means;

sign inversion means for generating, at a pixel position where no random number value is generated by said random number generation means, a value which has the same absolute value as but a different sign from a value output from said amplitude control means at an immediately preceding pixel position;

addition means for alternately adding the signal values generated by said amplitude control means and said sign inversion means to signal values input by said image input means; and

means for converting the signal values obtained by said addition means into n-values.

32. An image processing apparatus comprising:

image input means for inputting m-bit image information;

random number generation means for generating positive random number values at every other pixel positions;

random number sign control means for randomly converting the signs of the random number values generated by said random number generation means to positive or negative;

sign inversion means for generating, at a pixel position where no random number value is generated by said random number generation means, a value which has the same absolute value as but a different sign from a value output from said random number sign control means at an immediately preceding pixel position;

amplitude control means for alternately changing ratios of amplitudes of the signal values generated by said random number sign control means and said sign inversion means as a function of a signal value input by said image input means;

addition means for adding the signal values generated by said amplitude control means to signal values input by said image input means; and

means for converting the signal values obtained by said addition means into n-values.

33. An image processing apparatus comprising:

image input means for inputting m-bit image information;

random number generation means for generating positive or negative random number values at every other pixel positions;

amplitude control means for changing ratios of amplitudes of the random number values generated by said random number generation means as a function of a signal value input by said image input means;

sign inversion means for generating, at a pixel position where no random number value is generated by said random number generation means, a value which has the same absolute value as but a different sign from a value output from said amplitude control means at an immediately preceding pixel position;

addition means for alternately adding the signal values generated by said amplitude control means and said sign inversion means to signal values input by said image input means; and

means for converting the signal values obtained by said addition means into n-values.

34. An image processing apparatus comprising:

image input means for inputting m-bit image information;

random number generation means for generating positive or negative random number values at every other pixel positions;

sign inversion means for generating, at a pixel position where no random number value is generated by said random number generation means, a value which has the same absolute value as but a different sign from a value output from said random number generation means at an immediately preceding pixel position;

amplitude control means for alternately changing ratios of amplitudes of the signal values generated by said random number generation means and said sign inversion means as a function of a signal value input by said image input means;

addition means for adding the signal values generated by said amplitude control means to signal values input by said image input means; and

means for converting the signal values obtained by said addition means into n-values.

35. An image processing apparatus for converting m-bit input image information into n-bit data (m>n), and outputting the n-bit data, comprising:

random number signal generation means for generating random number signal values, maximum values of which being changed in accordance with the input image information at every "2.alpha.+1" pixel positions (.alpha.>0);

random number addition means for randomly adding positive and negative or negative and positive pairs of random number signal values generated by said random number signal generation means to the input image information at every a pixel position; and

image forming means for converting the image information obtained by said random number addition means into n-bit data.

36. The apparatus according to claim 35, wherein said image forming means comprises:

L-level pseudo halftoning means for respectively setting a quotient and a remainder obtained by dividing an m-bit value of the image information by an arbitrary constant to be upper- and lower-bit signals, binarizing the lower-bit signal by comparing the lower-bit signal with a random number, and adding the binarized lower-bit signal to the upper-bit signal; and

error correction means for error-correcting errors generated in error diffusion processing, said processing converting an L-level signal value obtained from the m-bit, M-level input image information by said pseudo halftoning means into the n-bit data, with respect to the m-bit, M-level input image information.

37. The apparatus according to claim 36, wherein said error correction means error-corrects a value obtained by performing a product-sum calculation on the errors generated in error diffusion processing which converts the L-level signal value obtained from the m-bit, M-level input image information by said pseudo halftoning means into the n-bit data, with weighting coefficients whose total sum total yields (M-1)/(L-1), with respect to the m-bit, M-level input image information.

38. The apparatus according to claim 35, wherein said image forming means comprises:

L-level pseudo halftoning means for respectively setting a quotient and a remainder obtained by dividing an m-bit value of the image information by an arbitrary constant to be upper- and lower-bit signals, binarizing the lower-bit signal by comparing the lower-bit signal with a dither signal, and adding the binarized lower-bit signal to the upper-bit signal; and

error correction means for error-correcting errors generated in error diffusion processing, said processing converting an L-level signal value obtained from the m-bit, M-level input image information by said pseudo halftoning means into the n-bit data, with respect to the m-bit, M-level input image information.

39. An image processing apparatus comprising:

input means for inputting an image;

detection means for detecting a density gradient amount of the image input by said input means;

pseudo random number generation means for generating a pseudo uniform random number a;

pseudo uniform random number applying means for applying the pseudo uniform random number a generated by said pseudo random number generation means to each pixel in an image region in which the density gradient amount detected by said detection means is larger than a predetermined threshold value, and applying a first positive value, a second negative value, and a third positive value based on the pseudo uniform random number a generated by said pseudo random number generation means, respectively, to each pixel in an image region in which the density gradient amount detected by said detection means is less than the predetermined threshold value; and

error diffusion processing means for performing error diffusion processing on the basis of the image to which the random numbers are applied by said pseudo uniform random number applying means.

40. The apparatus according to claim 39, wherein said detection means applies first and second filters for extracting direction features, which respectively correspond to a main scanning direction and a sub-scanning direction, to partial images of the image input by said input means so as to detect the density gradient amount of the image.

41. The apparatus according to claim 39, wherein the image input by said input means is an image obtained by performing predetermined density conversion with respect to an original image.

42. The apparatus according to claim 41, wherein the predetermined density conversion is log conversion.

43. The apparatus according to claim 39, wherein the first positive value, the second negative value, and the third positive value are respectively +a/2, -a, and +a/2.

44. The apparatus according to claim 39, wherein the application is an addition.

45. The apparatus according to claim 39, wherein image formation is performed on the basis of the image obtained by said error diffusion processing means.